Follow the Money: Compensation, Risk, and the Financial Crisis
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
STANFORD CLOSER LOOK SERIES
Topics, Issues, and Controversies in Corporate Governance and Leadership
Follow the Money: Compensation, Risk,
and the Financial Crisis
By David F. Larcker, Gaizka Ormazabal, Brian Tayan, and Daniel J. Taylor
September 8, 2014
introduction encourage excessive risk-taking with other people’s
It is commonly accepted by “Main Street” that Wall money.”5
Street bankers contributed to—if not caused out- How does compensation relate to and potential-
right—the financial crisis of 2008 by making large, ly dictate organizational risk? What is an “excessive”
leveraged bets on the housing market, and that risk and how is it distinguished from an acceptable
they were motivated to do so because of the large risk? Did the structure of executive compensation
bonuses they stood to receive if successful. To this contribute to the financial crisis? The answers to
end, a 2009 survey by KPMG finds that 52 percent these questions are valuable to policymakers as they
of senior managers at large financial institutions consider the means for preventing future crises and
believed that “incentives and remuneration” were to the directors of both bank and nonbank corpora-
most at fault in contributing to the credit crisis— tions as they consider how to design compensation
the highest of any attribute surveyed; 46 percent contracts that encourage appropriate but not exces-
of respondents reported that they were planning to sive risk-taking among employees.
review compensation policies in the wake of the cri-
sis.1 Similarly, a 2009 PricewaterhouseCoopers sur- Risk and Compensation
vey of financial services professionals finds that the Risk refers to the potential for loss due to a negative
three most frequently cited factors that created the outcome from an uncontrollable event. All corpo-
conditions for the crisis were a “culture and exces- rate activities—strategy, operations, investment—
sive risk-taking” (73 percent), “mispricing of risk” involve risk because their outcomes are uncertain.
(73 percent), and “rewards systems” (70 percent).2 For example, it is not practical to expect a financial
This line of reasoning has also been put forth by institution to eliminate all risk because in order to
prominent economists and policymakers. Accord- do so it would have to invest entirely in government
ing to former Federal Reserve Chairman Ben Ber- securities, guaranteeing a “risk-free” rate of return
nanke, “Compensation practices at some banking that is below its cost of capital. Rather than elimi-
organizations have led to misaligned incentives and nate risk entirely, each company must decide how
excess risk-taking, contributing to bank losses and much risk it is willing to assume given its expertise
financial instability.”3 In Congressional testimony, (its risk tolerance). Once established, the board de-
former Treasury Secretary Timothy Geithner ar- vises a compensation program that provides incen-
gued that, “Although many things caused this crisis, tive to management to pursue corporate objectives
what happened to compensation and incentives in in a manner consistent with this view of risk. In
creative risk-taking did contribute in some institu- this way, compensation not only encourages perfor-
tions to the vulnerability that we saw in this finan- mance but influences the manner in which finan-
cial crisis.”4 And economist and former Federal Re- cial results are achieved.
serve Vice Chairman Alan Blinder blamed the crisis Public commentators and governmental regu-
on “the perverse incentives built into the compen- lators often pay considerable attention to the
sation plans of many financial firms, incentives that incentives provided by an executive’s annual
stanford closer look series 1Follow the money: compensation, risk, and the financial crisis
compensation (typically referred to as flow pay). have shown that executives might decline to pur-
This includes the size of the current year’s overall sue new projects that would otherwise be valuable
pay package, the mix of cash and equity, the perfor- to well-diversified shareholders (i.e., projects with
mance measures that awards are contingent upon positive net present value) because they have more
(e.g., return-on-equity targets), and the time hori- at stake in the event of a loss than those sharehold-
zon over which they vest or are earned.6 The execu- ers.10 The executive’s risk tolerance becomes much
tive’s appetite for risk and the time horizon that he lower than that of shareholders, and over time, this
or she considers in making decisions will be influ- can reduce performance.
enced by the types of awards the board offers and Stock options can be used to address this prob-
the conditions under which they are earned. lem. The intrinsic value of stock options is a non-
However, an analysis of executive compensation linear function of share price. The value moves
that focuses exclusively on annual flow pay is incom- dollar-for-dollar with stock price when the option
plete because it fails to take into account the fact is “in the money” (when stock price is above exer-
that executives also have a large portfolio of person- cise price) but the value is unaffected by stock price
al wealth invested in the company. Many CEOs, when the option is “out of the money” (when the
particularly those who have been in their position stock price is below the exercise price). In other
for a number of years, accumulate a substantial in- words, as stock price falls below the exercise price
vestment in their companies by retaining vested eq- the executive is protected on the downside, but as
uity awards or by buying shares in their company.7 stock price rises above the exercise price the execu-
Over time, the incentive value of this portfolio will tive has unlimited potential upside. This introduces
begin to dominate the incentives provided by flow “convexity” into the executive’s potential payoff and
pay. In other words, executives will consider how encourages risk taking. Stock options tie the value
their decision making potentially affects their total of executive wealth to changes in stock price vola-
wealth rather than just one year’s pay. This is true of tility (measured by vega).11 As such, stock options
the CEOs of major banks, who hold considerable can be an effective tool to encourage managers to
amounts of company stock and options—both in become less risk-averse by investing in higher risk,
absolute terms and in relation to their annual sala- higher return investments. Research shows that
ries. For example, in 2006 (prior to the financial executives understand that the expected value of
crisis), the average CEO of a bank included in the a stock option increases with the volatility of the
S&P 1500 Index was paid $5.8 million but held stock price and that executives tend to respond to
$106 million of stock and options.8 stock option awards by investing in riskier projects.
So how does this portfolio of stock and options For example, Coles, Daniel, and Naveen (2006)
influence risk taking? It depends on the composi- find that executives with large stock option expo-
tion. An executive whose wealth consists entirely of sure (high vega) spend more money on research
direct stock investments—either restricted shares or and development, reduce firm diversification, and
shares that have vested but not been sold—stands increase firm leverage—all actions that increase the
to gain or lose wealth dollar-for-dollar with chang- risk profile of the firm.12 In the banking industry,
es in the stock price (measured by delta).9 Many Mehran and Rosenberg (2007) find that large stock
boards like this arrangement because it is seen as option exposure is positively correlated with firm
putting the executive on equal footing with the av- risk, measured by asset and equity volatility.13
erage investor. However, this is not entirely the case What incentives did the CEOs of financial insti-
because the average investor holds shares as part of tutions have to take risk prior to the financial crisis?
a diversified portfolio, while the risk-averse execu- Exhibit 1 shows the vega of CEO wealth over the
tive typically has a large, concentrated exposure to 18 year period 1992 to 2009. It shows that equi-
a single stock. Concentrated exposure to a single ty-based risk-taking incentives were consistently
stock exposes executives to greater risk than is ex- higher among banks than nonbank corporations,
perienced by diversified shareholders. Researchers and consistently highest among the subset of banks
stanford closer look series 2Follow the money: compensation, risk, and the financial crisis
that originated and distributed securitized assets.14 incentives they provide to CEOs by replacing
Notably, there is a dramatic increase in risk-taking stock option grants with direct equity awards.
incentives in 2000 following deregulation of the How much risk should bank executives be en-
banking industry and the repeal of Glass-Steagall couraged to take?
(which paved the way for an aggressive expansion of 4. One approach to reducing risk in the banking
banks’ securitization activity).15 By 2006, the vega of industry is for regulators to monitor the riskiness
the average securitizing bank CEO was fifteen-fold of bank assets and restrict the amount of leverage
higher than it had been in 1992 and quadruple that that banks can take (e.g., through stress tests or
of the average nonbank CEO. Thus the incentives fixed limits on leverage). Another approach is for
that bankers had to take risk not only increased but boards to restructure compensation contracts to
increased substantially in the years preceding the reduce the incentives that executives have to take
crisis, subsequent to the repeal of Glass-Steagall. risk in the first place. What are the advantages
Because equity-based incentives are positively to regulatory oversight of bank activity, versus
correlated with corporate risk, the immediate ques- changes to the incentive structure? If regulators
tion for the board of directors is whether these monitor bank activity but the incentives for risk-
incentives induce “excessive” risk taking. Unfortu- taking are still present, will managers find new
nately, no standard litmus test exists to distinguish ways to “game the system?”
an excessive risk from acceptable risk. An excessive
risk might be one whose downside is so large that 1
KPMG, “Never Again? Risk Management in Banking beyond the
Credit Crisis” (2009).
the firm cannot financially bear it.16 2
PricewaterhouseCoopers and Economist Intelligence Unit, “Re-
To this end, much public attention has been ward: A New Paradigm?” (September 2008).
paid to whether certain corporations are “too big to 3
Cited in: Federal Reserve Press Release (October 22, 2009).
4
Timothy Geithner, “Testimony to Senate Appropriations Subcom-
fail,” meaning that the government and taxpayers mittee on the Treasury Department’s Budget Request,” Reuters (June
bear the downside and not just the firm’s sharehold- 6, 2009).
5
Edited lightly for clarity. Alan S. Blinder, “Crazy Compensation and
ers. In this situation, “excessive” risk taking from
the Crisis,” The Wall Street Journal (May 28, 2009).
the view point of regulators might be different from 6
This information is easy to observe because it is included in public
that of the board and shareholders. filings (form DEF 14A with the SEC) and often summarized by
business reporters and consulting firms.
7
Furthermore, many companies adopt equity ownership guidelines
Why This Matters that require executives to own a minimum amount of company
stock. According to Equilar, approximately 84 percent of large com-
1. Compensation committees put considerable ef-
panies currently have such guidelines. While ownership require-
fort into designing CEO pay packages that en- ments vary significantly across companies, a typical plan requires the
courage performance and tie compensation to CEO to hold approximately five times his or her annual salary in
equity, demonstrating the great extent to which equity ownership
results. How well do boards understand the rela- dwarfs flow pay among most companies. See: Equilar, “Executive
tion between compensation and risk? Stock Ownership Guidelines” (2013).
8
Data provided by Standard and Poor’s ExecuComp. “Banks” refers
2. Shareholders and the public are intensely fo-
to bank holding companies that file Call reports with the Federal Re-
cused on the issue of CEO pay. However, the serve and appear on Standard & Poor’s ExecuComp (i.e. appear on
size and composition of CEO’s equity wealth is the S&P 1500). CEO pay is comprised of salary, bonus, total value
of restricted stock granted that year, total value of options granted
often a greater source of incentive than annual that year, long-term incentive payouts, and “all other compensation”
flow pay. How much attention do directors pay listed in the firm’s Summary Compensation Table. Value of equity
securities calculated according to Core and Guay (2002). See John
to the risk-taking incentives provided by CEO
E. Core and Wayne R. Guay, “Estimating the Value of Employee
wealth? Do boards evaluate the relation between Stock Option Portfolios and Their Sensitivities to Price and Volatil-
the CEO’s personal risk incentives and the over- ity,” Journal of Accounting Research (2002).
9
Delta is calculated as the ratio of the change in expected value of the
all risk tolerance of the firm? executive’s total portfolio to the change in underlying share price,
3. A long list of regulations have been enacted or and is conventionally reported using a 1 percent change in stock
price.
proposed to reshape the banking industry fol- 10
See Clifford W. Smith and René M. Stulz, “The Determinants of
lowing the crisis. However, a simple solution Firms’ Hedging Policies,” Journal of Financial and Quantitative
might be for directors to lower the risk-taking Analysis (1985); Richard A. Lambert, David F. Larcker, and Robert
stanford closer look series 3Follow the money: compensation, risk, and the financial crisis
E. Verrecchia, “Portfolio Considerations in Valuing Executive Com- research assistance in the preparation of these materials.
pensation,” Journal of Accounting Research (1991); and Stephen A.
Ross, “Compensation, Incentives, and the Duality of Risk Aversion The Stanford Closer Look Series is a collection of short
and Riskiness,” Journal of Finance (2004).
case studies that explore topics, issues, and controver-
11
Vega is calculated as the ratio of the change in expected value of the
executive’s total portfolio to the change in volatility of the underly- sies in corporate governance and leadership. The Closer
ing share price, and is conventionally reported using a change in Look Series is published by the Corporate Governance
standard deviation of 0.01. Research Initiative at the Stanford Graduate School
12
Jeff L. Coles, Naveen D. Daniel, and Lalitha Naveen, “Managerial of Business and the Rock Center for Corporate Gover-
Incentives and Risk-Taking,” Journal of Financial Economics (2006).
nance at Stanford University. For more information, visit:
See also: Christopher S. Armstrong, David F. Larcker, Gaizka
Ormazabal, and Daniel J. Taylor, “The Relation between Equity http://www.gsb.stanford.edu/cldr.
Incentives and Misreporting: The Role of Risk-taking Incentives,” Copyright © 2014 by the Board of Trustees of the Leland
Journal of Financial Economics (2013).
Stanford Junior University. All rights reserved.
13
Hamid Mehran and Joshua Rosenberg, “The Effect of CEO Stock
Options on Bank Investment Choices, Borrowing, and Capital,”
Federal Reserve Bank of New York Staff Report No. 305 (October
2007, revised June 2008).
14
During the financial crisis the risk of the entire banking sector in-
creased. It turns out that this sector-wide increase in risk coincided
with a massive sector-wide increase in risk-taking incentives. This
increase was particularly pronounced among banks that were partici-
pating in the securitization process. Prior research suggests that the
increasingly common practice of securitizing loans led to a deteriora-
tion in loan quality and was at the heart of the financial crisis. A large
increase in risk-taking incentives among securitizing banks could po-
tentially explain the deterioration in loan quality at these banks. See:
Benjamin J. Keys, Tanmoy Mukherjee, Amit Seru, and Vikrant Vig,
“Did Securitization Lead to Lax Screening: Evidence from Subprime
Loans,” Quarterly Journal of Economics (2010).; Benjamin J. Keys,
Amit Seru, and Vikrant Vig, “Lender Screening and the Role of
Securitization: Evidence from Prime and Subprime Mortgage Mar-
kets,” Review of Financial Studies (2012); and Atif Mian and Amir
Sufi, “The Consequences of Mortgage Credit Expansion: Evidence
from the US Mortgage Default Crisis,” The Quarterly Journal of Eco-
nomics (2009).
15
In November 1999, the Gramm-Leach-Bliley Financial Moderniza-
tion Act repealed Section 20 of the Glass-Steagall Act. Prior research
argues that deregulation of the banking industry led to a more com-
petitive environment, demand for talent, and an increase in equity-
and option-based compensation. See R. Glenn Hubbard and Darius
Palia, “Executive Pay and Performance: Evidence from the U.S.
Banking Industry,” Journal of Financial Economics (1995).
16
Risk is measured by the probability of gain/loss prior to an event
taking place. A negative outcome does not tell us that an activity was
highly risky any more than a positive outcome tells us that an activity
was not risky. Still, the inability of the financial services industry to
absorb the losses it incurred certainly suggests that risk taking was
excessive in hindsight.
David Larcker is Director of the Corporate Governance
Research Initiative at the Stanford Graduate School of
Business and senior faculty member at the Rock Center
for Corporate Governance at Stanford University. Gaizka
Ormazabal is Assistant Professor of Accounting and Con-
trol at IESE Business School at the University of Navarra.
Brian Tayan is a researcher with Stanford’s Corporate Gov-
ernance Research Initiative. Daniel Taylor is Assistant Pro-
fessor of Accounting at Wharton. Larcker and Tayan are
coauthors of the books A Real Look at Real World Cor-
porate Governance and Corporate Governance Matters.
The authors would like to thank Michelle E. Gutman for
stanford closer look series 4Follow the money: compensation, risk, and the financial crisis
Exhibit 1 — risk-taking incentives prior to the financial crisis
500
Repeal of Glass Steagall Crisis
400
300
200
100
0
1992 1994 1996 1998 2000 2002 2004 2006 2008
Nonbanks Banks Securitizing Banks
Note: Plot of average portfolio vega by year for all CEOs on ExecuComp with non-missing data. Portfolio vega (in thou-
sands) is calculated as the sensitivity of the CEO’s equity portfolio to a 0.01 change in stock volatility per Core and Guay
(2002) and is expressed in units of $1,000. The term “Banks” refers to bank holding companies (132 unique banks), “Securi-
tizing Banks” refers to “Banks” with at least six quarters of non-zero securitized assets (58 unique banks), and “Nonbanks”
refers to all other firms (3,232 unique firms). Section 20 of the Glass-Steagall Act was repealed by the Gramm-Leach-Bliley
Financial Modernization Act in November 1999; effective March 2000, bank holding companies were allowed to expand
the issuance of asset-backed securities.
Source: Data from ExecuComp. Calculations by the authors.
stanford closer look series 5You can also read